There is a drastic shortage of transplantable livers in the US with approximately 17,000 individuals each year on the waiting list, but only 6,000 per year receiving transplants. This shortage can be significantly reduced if marginal livers with suboptimal characteristics, such as those with prolonged warm ischemia time (WIT) due to cardiac death donor (DCD), older donor age, or moderate steatosis, can also be included in the pool of transplantable livers. In this project, we propose subnormothermic machine perfusion (SNMP) as a platform to metabolically precondition and recover marginal livers prior to transplantation, thus expanding the donor pool size. In Aim 1 of this proposal, our objective is to test SNMP for the first time as a proof of concept on discarded human livers, which get donated to our lab approximately once per week. We will collect perfusate, bile, and tissue biopsy samples to analyze the liver's dynamic metabolic profile in perfusion. The tissue biopsy samples will be sent to the UC Davis Metabolomics Core to determine relative levels of ~150 intracellular metabolites from primary metabolites and ~300 lipid compounds. We will then correlate the dynamic metabolic profiles to donor characteristics using multi-way principal component analysis and multiple regression analysis. The expected outcome of this aim will be to develop a new set of cutoff metrics based on donor characteristics (age, WIT, and steatosis) to determine if a liver is transplantable after SNMP. In Aim 2, our objective is to better understand metabolic dynamics during perfusion and gain more mechanistic insight into how SNMP reverses WIT-induced injury. We will borrow computational modeling approaches from systems biology to explain trends observed with the time-course intracellular metabolomics data and determine which metabolic pathways are up or down regulated during perfusion. We will construct a stoichiometric/regulatory reaction network to describe hepatocyte metabolism, and apply structural kinetic modeling (SKM) to determine the most likely explanation of pathway flux changes to explain the metabolite data. By comparing the dynamics of a healthy liver, and one with long WIT, our expected outcome of this aim is to discover which metabolic pathways are responsible for reversing ischemic injury in perfusion. Our long term goal is to replace the current system of donor liver assessment to include an SNMP step, which will provide two key benefits: 1.) Improve the metabolic condition of suboptimal grafts, particularly those from DCDs with longer WITs and 2.) Provide the transplant surgeon in the operating room with a more robust set of quantitative biomarkers based on perfusion measurements to assess if a liver can be transplanted. The broader impact of this work is that it has the potential to save thousands of lives of those with liver failure. In addition, this study is, to our knowledge, the first to integate systems biology and metabolomics data to model the dynamics of a live human organ and should serve as an example for others trying to use time-course biopsies to model tissue metabolism in vivo.